Sign Up to like & get
recommendations!
1
Published in 2022 at "IEEE Access"
DOI: 10.1109/access.2022.3231455
Abstract: We propose the position-based scaled gradient (PSG) that scales the gradient depending on the position of a weight vector to make it more compression-friendly. First, we theoretically show that applying PSG to the standard gradient…
read more here.
Keywords:
gradient;
model;
compression;
position based ... See more keywords
Photo from wikipedia
Sign Up to like & get
recommendations!
0
Published in 2021 at "IEEE Computational Intelligence Magazine"
DOI: 10.1109/mci.2021.3084393
Abstract: While deep neural networks (DNNs) deliver state-of-the-art accuracy on various applications from face recognition to language translation, it comes at the cost of high computational and space complexity, hindering their deployment on edge devices. To…
read more here.
Keywords:
deep neural;
neural networks;
model;
energy ... See more keywords
Sign Up to like & get
recommendations!
0
Published in 2020 at "IEEE Transactions on Computers"
DOI: 10.1109/tc.2020.2970917
Abstract: Modern distributed engines are increasingly deployed to accelerate large-scaled deep learning (DL) training jobs. While the parallelism of distributed workers/nodes promises the scalability, the computation and communication overheads of the underlying iterative solving algorithms, e.g.,…
read more here.
Keywords:
learning systems;
critical set;
model;
model compression ... See more keywords
Sign Up to like & get
recommendations!
2
Published in 2022 at "IEEE Transactions on Circuits and Systems for Video Technology"
DOI: 10.1109/tcsvt.2022.3197395
Abstract: The existing end-to-end optimized 3D action recognition methods often suffer from high computational costs. Observing that different frames and different points in point cloud sequences often have different importance values for the 3D action recognition…
read more here.
Keywords:
model compression;
model;
action recognition;
Photo from wikipedia
Sign Up to like & get
recommendations!
1
Published in 2022 at "IEEE transactions on neural networks and learning systems"
DOI: 10.1109/tnnls.2022.3165123
Abstract: Although neural networks have achieved great success in various fields, applications on mobile devices are limited by the computational and storage costs required for large models. The model compression (neural network pruning) technology can significantly…
read more here.
Keywords:
channel pruning;
network channel;
differentiable network;
model compression ... See more keywords
Sign Up to like & get
recommendations!
2
Published in 2023 at "IEEE transactions on neural networks and learning systems"
DOI: 10.1109/tnnls.2023.3266435
Abstract: Observing that the existing model compression approaches only focus on reducing the redundancies in convolutional neural networks (CNNs) along one particular dimension (e.g., the channel or spatial or temporal dimension), in this work, we propose…
read more here.
Keywords:
multidimensional pruning;
model compression;
dimension;
mdp ... See more keywords
Sign Up to like & get
recommendations!
1
Published in 2021 at "Future Internet"
DOI: 10.3390/fi13120300
Abstract: Despite the advance in deep learning technology, assuring the robustness of deep neural networks (DNNs) is challenging and necessary in safety-critical environments, including automobiles, IoT devices in smart factories, and medical devices, to name a…
read more here.
Keywords:
adversarial training;
manifold adversarial;
model;
model compression ... See more keywords